[GRASS-dev] testing the new network modules

Moritz Lennert mlennert at club.worldonline.be
Thu Nov 26 04:14:26 EST 2009


Dear Daniel,

First of all thanks once again for all the new network modules, they are
a great addition.

I haven't found as much time as I would like to to test them, but here 
are first remarks:

- v.net.distance: this module needs at least a min threshold, and
ideally a max threshold (just like v.distance). As an exemple, I'm
trying to find for each of over 1200 cities in Europe the closest other
city. Currently, when I launch

v.net.distance input=transtools_roads_fuas at transtools
output=fuas_dist_time_network from_layer=2 from_cats=3-1250 to_layer=2
to_cats=3-1250 afcolumn=temps

I always get cat=tcat and dist=0, but I would like to exclude the
closest city (i.e. min>0). If I see correctly, however, you have
"hardcoded" the algorithm in the library functions, so this would
probably need some rearrangement in order to allow the module to alter
the dist thresholds. Currently, to get my distances to the closest other
city, I need to loop through all combinations and use v.net.path, and
then filter for the shortest distance.


- In addition, it would be great if v.net.distance provided a function
similar to '-a' in v.distance, i.e. the construction of a complete
distance matrix.


I'll come back to you on other issues once I've found the time...

Moritz


More information about the grass-dev mailing list